The thesis approaches the problem of delusion as delusion occurs in the clinical field, and identifies two questions to be addressed. The first question is: which account of delusion most comprehensively addresses the dataset of clinical features of delusion? The second question asks: is this successful account an empiricist or a rationalist or a different approach?
Three delusions are chosen as test cases because together they represent a wide range of the data set of clinical features. The delusions are: erotomanic delusion; Capgras delusion; and perceptual delusional bicephaly.
First the prevailing theories of delusion are examined. These are: the empiricist one-factor approach; the empiricist two-factor approach; and the (relationist) rationalist approach to delusion. The conclusion is that in virtue of stipulations which each of the prevailing theories variously makes concerning neuropathology; (deep) irrationality; and abnormal experience, all of the accounts encounter substantial problems. I use this conclusion as warrant to propose an alternative account.
The alternative account which I develop combines two discrete theories. The theories are: Daniel Dennett’s approach to mind and Prediction Error theory as applied to the brain. First I build up a combined Dennettian plus Prediction Error (D-PE) model. This is then applied to the test cases of delusion, where it is shown to comprehensively meet the features of each case.
Therefore to answer the first thesis question, a D-PE approach affords the greatest explanatory potential. And to answer the second thesis question, while a D-PE account is technically a rationalist approach to delusion, it is different because, pace rationalism as typically construed, the D-PE model eschews the idea of (deep) irrationality and it permits abnormal experience an explanatory role. Therefore a D-PE approach adds to the ongoing conversation by offering a different way of thinking about the problem of delusion in the clinical field.
